Project Summary Gait dysfunction is notoriously difficult to rehabilitate and, despite recent advances, we lack effective treatment options for many clinical populations. Here we propose experiments that will establish a new way of thinking about gait rehabilitation after stroke. Instead of training a patient to perform a particular walking pattern repetitively (as is often done in conventional therapy), we suggest that rehabilitation should aim to make the desired walking pattern cost less energy because people naturally prefer walking patterns that minimize energy cost. With this in mind, we will execute a series of experiments that demonstrate that persons post-stroke are 1) capable of manipulating the symmetry of their walking patterns, and 2) adopt new gait patterns if the new patterns cost less energy. These introductory studies will lay an important foundation for designing gait rehabilitation interventions focused on decreasing the energy cost of desired gait patterns. In Aim 1, we will investigate the capacity for persons post-stroke to change gait symmetry. Many persons post- stroke walk with asymmetric steps in daily life. We have designed a visual feedback system where we can show participants how symmetrically they are walking in terms of either step position (i.e., where the feet are placed in global space) or step length (i.e., how far one foot is placed ahead of the other). We hypothesize that persons post-stroke will be able to use the visual feedback to improve their gait symmetry in both step position and step length, revealing that persons post-stroke retain the capacity to walk more symmetrically than they do in daily life. This leads to an important question that we will address in Aim 2 ? if persons post-stroke can walk more symmetrically, why do they prefer asymmetric walking patterns? In Aim 2, we will investigate how energy cost influences walking after stroke. Persons post-stroke often walk with asymmetric steps and the energy cost of walking is elevated relative to healthy adults. However, it is possible that persons post-stroke prefer to walk asymmetrically because symmetric stepping is even more effortful given the unilateral motor deficits that affect many patients. We will assess how speed and asymmetry affect the energy cost of walking in persons post-stroke by having participants walk at different speeds and manipulate their asymmetry using visual feedback. We hypothesize that persons post-stroke will not experience any energetic benefit from walking symmetrically. We will then use a novel treadmill controller to create an environment where symmetric walking costs less energy than asymmetric walking. We hypothesize that persons post-stroke will walk more symmetrically in this environment, revealing that persons post-stroke may naturally walk more symmetrically if rehabilitation can reduce the energy cost of symmetric walking. Overall, this proposal will establish that persons post-stroke can manipulate their gait symmetry and will change how they walk to minimize energy cost. These findings will establish a foundation for building new rehabilitation interventions that drive improvements in patient gait patterns by reshaping energetic landscapes.